I have a scene where I’m populating my scene using a single Redshift proxy that has been duplicated a bunch of times. However when I send it to render on Deadline, some of the worker nodes fail the task and go on to the next one or they just sit on the task at 0% indefinitely.
I’ve render the same scene with the matrix object that is duplicating the RS Proxy turned off and it renders without issue.
Having never used RS Proxies before, does Deadline or Worker have a hard time rendering loads of duplicated proxies?
Screenshot of Job Report: Screenshot - 66ccfa129dfcf54ede09577ea94b77ce - Gyazo
From that screenshot all we know is C4D is crashing without saying why, so we’ll need to run C4D outside of the C4DBatch application plugin.
The issue with our C4DBatch plugin is we start C4D and talk to it over a socket, so in cases like these we don’t get to see the final messages from a crashing application.
So what we’ll do is go through the troubleshooting guide, the first step of which is to re-submit the job with ‘Use Batch Plugin’ unchecked. That might make logs that you can use to see what’s wrong, but if not keep going through that guide.
As for the machines that sit at 0% indefinitely that’s usually because the C4D Commandline license hasn’t been set up on that machine yet and it’s sitting there waiting for you to configure it. Check the Worker’s log to see if it matches up with this Maxon article on licensing issues.
Ah okay gotcha.
I’ve ran the troubleshooting guide and sent over what command prompt said.
As for the 0%, I’ve double checked all the licencing on the render slaves and they’re all valid. Even Commandline C4D doesn’t throw up any errors on that front.
As for the crashing issue, that only happens on the same file I submit and that file just so happens to have the Redshift Proxies in it. All other jobs don’t throw up such an error.